Recursive Estimation in Non-Linear Time Series Models of Autoregressive Type
Knut Kristian Aase · Journal of the Royal Statistical Society Series B (Statistical Methodology) · 1983
SUMMARY An adaptive estimation principle is considered for non-standard time series models; non-linear autoregressive processes in one dimension. Because of the recursive nature of the resulting estimator, it is computationally appealing, especially when a time series is considered as a flow of data. The least squares estimator is, for example, quite likely to be inferior. Asymptotic properties are derived under fairly weak assumptions, and special applications are considered. Monte Carlo simulations are used for illustration.